BLAST qcov equals to 0

Question

[Ming]

Hello All,

I just performed a blastn, and I have some alignment returning qcovs = 0. I am not sure how this is possible because the alignment reports a pident and a bitscore for that alignment. For instance:

    qseqid sseqid pident bitscore evalue qcovs qcovhsp length
k127_1554010 ref|NZ_AONQ01000017.1| 77.007 409 2.85e-110 0 0

From my understanding, qcovs = query coverage. A qcovs = 0 means that 0% of the query is aligned to the subject. But why is the pident and bitscore reported as positive values?

Thank you in advanced!

[mgdias.jose]

Can you provide the parameters that you have defined for your blast?

[lieven.sterck]

that output is not corresponding to the header you provide, length has no value (from the top of my head, I even think 'length' is not a valid field)

can you provide your actual blast command (as also mgdias.jose requested)

Answer 1

[h.mon]

As lieven.sterck noted above, the output does not correspond to the header - which may indicate you did something wrong / the output fields are not exactly what you think they are.

I will have an attempt at another (rather wild) guess:

The name of your query and subject indicates you are blasting MEGAHIT contigs against bacterial genomes - potentially very long sequences. If you have a strong short similarity, blast will output the hit, but qcov may be set to zero because it is a tiny fraction of a percentage. Try to use qstart, qend, sstart and send to see how long are the alignments.